Object detection device

Abstract

Object detection device installed in a vehicle for detecting an object located outside the vehicle, comprising: a transmitting unit (1, 2) configured to transmit probe waves which are ultrasonic waves whose frequency changes over time in a predetermined pattern, a receiving unit (1, 5) configured to receive ultrasound waves, a matching calculation unit (11) configured to calculate a matching ratio between a frequency of a received wave, which is an ultrasonic wave received by the receiving unit, and the specified pattern, an amplitude maximum detection unit (10) configured to detect a maximum of the amplitude of the received wave, and a distance determination unit (4) configured to determine a distance from the vehicle to the object based on the degree of matching calculated by the degree of matching calculation unit and a result of a detection of a maximum of the amplitude of the received wave obtained by the amplitude maximum detection unit.

Description

Technical field

[0001] The present invention relates to an object detection device. State of the art

[0002] An object detection device, intended to be installed in a vehicle, for detecting obstacles by sending or transmitting and receiving ultrasonic waves, as disclosed in EP 2 373 434 B1, uses a known method which varies the frequency of a probe wave over time and compares the frequency of the probe wave with frequencies of received waves in order to avoid interference of ultrasonic waves emitted by the vehicle having the device and other vehicles driving around the vehicle having the device.

[0003] The object detection device disclosed in EP 2 373 434 B1, however, uses a resonant microphone to transmit and receive ultrasonic waves. In such an object detection device, the frequency waveform of a received signal can vary in the same or a similar way to the frequency waveform of the probe wave at the start and end times of reception of the reflected wave. A method for identifying received waves that uses only a frequency comparison can therefore be subject to a reduction in object detection accuracy.

[0004] From DE 10 2010 033 384 A1 a method and a device for evaluating an echo signal for vehicle environment detection as well as an associated device for vehicle environment detection are also known.

[0005] The object of the present invention is to provide an object detection device that increases the accuracy of object detection.

[0006] The problem is solved by the subject matter of the main claim. Advantageous further developments are specified in the dependent claims.

[0007] According to the invention, if the amplitude of the reflected wave is relatively large, a frequency pattern present in the probe wave will also appear in the reflected wave. Consequently, determining the distance from the vehicle to the object based on the degree of match between a frequency and the detection of a maximum amplitude can improve the accuracy of object detection. Brief description of the characters

[0008] They show: Fig. 1 a block diagram of an object detection device according to a first embodiment, Fig. 2. An example of frequency changes of a positive chirp signal, Fig. 3. An example of frequency changes of a negative chirp signal, Fig. 4 Another example of frequency changes of a positive chirp signal, Fig. 5 a representation of an amplitude threshold and a reception time, Fig. 6. A representation of calculating a degree of compatibility using a frequency offset. Fig. 7 a flowchart of object acquisition processing in the first embodiment, Fig. 8 a flowchart of an amplitude maximum detection processing in the first embodiment, Fig. 9 a presentation of advantages of the first embodiment, Fig. 10 a flowchart of object acquisition processing according to a second embodiment, Fig. 11 a representation of a calculation of a residual sum of squares according to a third embodiment, Fig. 12 a flowchart of an amplitude maximum detection processing in the third embodiment, Fig. 13 a representation of a calculation of a residual sum of squares according to a fourth embodiment, Fig. 14 a flowchart of an amplitude maximum detection processing in the fourth embodiment, Fig. 15 a block diagram of an object detection device according to a fifth embodiment, Fig. 16 an example of frequency changes of a chirp signal into another embodiment, Fig. 17 an example of frequency changes of a chirp signal in another embodiment, Fig. 18 an example of frequency changes of a chirp signal in another embodiment, Fig. 19 an example of frequency changes of a chirp signal in another embodiment, Fig. 20 an example of frequency changes of a chirp signal in another embodiment, Fig. 21 an example of frequency changes of a chirp signal in another embodiment, and Fig. 22 an example of frequency changes of a chirp signal in another embodiment. Description of specific embodiments

[0009] Several embodiments of the present invention are described below with reference to the accompanying figures. The same reference numerals are assigned to essentially the same elements or steps in the embodiments, and any redundant description thereof is omitted. First embodiment

[0010] A first embodiment is now described. An object detection device of the present embodiment is an ultrasonic device. The object detection device is mounted on a vehicle and detects an object that is located outside the vehicle.

[0011] The object detection device, as shown in Fig. Figure 1 shows a microphone 1, a transmitter circuit 2, a signal generation unit 3, and a control unit 4. The object detection device further comprises a receiver circuit 5, a signal processing unit 6, an amplitude generation unit 7, a frequency generation unit 8, an amplitude threshold determination unit 9, an amplitude maximum detection unit 10, a frequency determination unit 11, and a reference wave memory 12.

[0012] The control unit 4, the signal processing unit 6, and the like are configured as a known microcomputer comprising a central processing unit (CPU), a read-only memory (ROM), a random-access memory (RAM), an input / output unit (I / O unit), and the like, and performing various types of processing, such as arithmetic processing, according to programs stored in the ROM or the like. The control unit 4 or the like may be configured as an application-specific integrated circuit (ASIC) comprising a signal processing circuit.

[0013] Microphone 1 is mounted on an external surface of the vehicle and transmits ultrasonic waves, which are probe waves, to the outside of the vehicle to detect objects. Microphone 1 includes, in particular, a piezoelectric element (not shown) that provides a piezoelectric layer between two opposing electrodes. The two electrodes are connected to the transmitting circuit 2. When an alternating current (AC) voltage is applied by the transmitting circuit 2, and the piezoelectric layer deforms or changes its shape, an ultrasonic wave is transmitted from microphone 1 towards the outside of the vehicle.

[0014] Transmitter circuit 2 amplifies an input signal and outputs the amplified signal. Signal generator unit 3, which generates a pulse signal, is connected to transmitter circuit 2. Transmitter circuit 2 amplifies a pulse signal input from signal generator unit 3 into an AC signal and applies an AC voltage to microphone 1.

[0015] After receiving a pulse signal, which is generated as an AC signal by the signal generation unit 3, the microphone 1 and the transmitter circuit 2 transmit a probe wave whose frequency corresponds to a frequency of the pulse signal. The microphone 1 and the transmitter circuit 2 constitute a transmitter unit.

[0016] In response to a transmit instruction from the control unit 4, the signal generator 3 generates a pulse signal containing a chirp signal whose frequency changes over time in a predefined pattern. This results in the transmission of a probe wave, an ultrasonic wave containing a chirp signal whose frequency changes over time in a predefined pattern, from the microphone 1. The microphone 1 is configured to transmit probe waves with a variety of patterns, each containing different types of chirp signals. A probe wave pattern is determined according to the transmit instruction sent from the control unit 4 to the signal generator 3. To ensure a constant probe wave amplitude, the signal generator 3 and the transmit circuit 2 are controlled to maintain constant amplitudes in both the pulse signal and the AC voltage input to the microphone 1.

[0017] In the present embodiment, a chirp signal whose frequency increases monotonically over time and a chirp signal whose frequency decreases monotonically over time are used.

[0018] In particular, when a transmission instruction for a first pattern is sent from the control unit 4 to the signal generation unit 3, the signal generation unit 3 generates a pulse signal whose frequency, as in Fig. As shown in Figure 2, the frequency increases linearly over time. Thus, a first probe wave, which exhibits a positive chirp signal or up-chirp signal whose frequency increases over time, is transmitted by microphone 1.

[0019] In particular, when a transmission instruction for a second pattern is sent from the control unit 4 to the signal generation unit 3, the signal generation unit 3 generates a pulse signal whose frequency changes over time, as shown in Fig. Figure 3 shows a linear decrease. Thus, a second probe wave, which exhibits a negative chirp signal or down-chirp signal whose frequency decreases over time, is transmitted by microphone 1.

[0020] Assuming that f0 represents a resonant frequency of microphone 1, the signal generator 3 begins to cycle through pulse signal frequencies from a frequency different from the resonant frequency f0. Specifically, the signal generator 3 cycles through pulse signal frequencies between a frequency lower than the resonant frequency f0 and a frequency higher than the resonant frequency f0. The frequency of the pulse signal generated by the signal generator 3 and the frequency of the probe wave can be continuous or, as shown in Fig. 4 shown, change discretely.

[0021] Microphone 1 is configured to transmit and receive an ultrasonic wave and output a voltage corresponding to the sound pressure level of the received ultrasonic wave. The two electrodes of the piezoelectric element within microphone 1 are also connected to the receiver circuit 5. Accordingly, a voltage applied between the two electrodes when microphone 1 receives an ultrasonic wave and the piezoelectric layer changes shape is input to the receiver circuit 5. The receiver circuit 5 uses an amplifier (not shown) to amplify the weak voltage input from microphone 1 to a desired voltage level, a bandpass filter to remove noise as needed, and an analog-to-digital converter (ADC) to convert the voltage input from microphone 1 into a digital signal for output.

[0022] Microphone 1 and receiver circuit 5 are thus configured to receive an ultrasonic wave and output a signal corresponding to the amplitude and frequency of the received ultrasonic wave. Microphone 1 and receiver circuit 5 constitute a receiver unit, and the ultrasonic wave received by microphone 1 is a received wave.

[0023] The signal processing unit 6 detects a frequency and amplitude of the received wave using a signal obtained through A / D conversion by the receiver circuit 5.

[0024] The signal processing unit 6 specifically calculates components I and Q of a mixed signal consisting of a reference signal generated by the signal generation unit 3 and the received signal, using orthogonal demodulation. It is assumed that t represents time and f prepresents a frequency of the reference signal, I represents a magnitude of a signal obtained by multiplying an output signal of the receiving circuit 5 by sin(2πf p t) and then removing components whose frequency is 2f p or greater. Q represents the magnitude of a signal obtained by multiplying the output signal of the receiving circuit 5 by cos(2πf). p t) and then removing a component whose frequency is 2f p or greater, is obtained. The frequency f p The frequency f takes on a value close to the resonant frequency f0 within a resonant bandwidth of microphone 1. p can be set to f0.

[0025] The amplitude generation unit 7 calculates an amplitude A r the received wave according to A r = (I 2 + Q 2 ) 1 / 2 The frequency generation unit 8 calculates a frequency f rthe received wave according to f r = 1 / (2π)·dP / dt + f p , where P represents a phase P of the received wave, which is given by P = atan (Q / I).

[0026] The amplitude generator unit 7 generates an amplitude waveform based on the calculated amplitude Ar. The frequency generator unit 8 generates a frequency waveform based on the calculated frequency fr. The amplitude waveform generated by the amplitude generator unit 7 is to be sent to the amplitude threshold determination unit 9 and the amplitude maximum detection unit 10. The frequency waveform generated by the frequency generator unit 8 is to be sent to the frequency determination unit 11.

[0027] The amplitude threshold determination unit 9 determines whether the amplitude of the received wave is greater than a predefined amplitude threshold, based on the amplitude waveform generated by the amplitude generation unit 7. The result of a determination made by the amplitude threshold determination unit 9 is sent to the control unit 4. The amplitude threshold can be constant over time or it can vary with the passage of time since the transmission of the probe wave. The amplitude threshold can, for example, be, as in Fig. As shown in section 5, the effect decreases discreetly over time.

[0028] The propagation time T from the time at which microphone 1 transmits a probe wave (hereinafter referred to as the transmission time) to the time at which microphone 1 receives its reflected wave (hereinafter referred to as the reception time) is given by T = 2D / c. Here, D represents the distance from microphone 1 to an object that has reflected the probe wave, and c represents the speed of sound. A predetermined threshold for the propagation time T is set in the object detection device. The amplitude threshold determination unit 9 detects a reflected wave by comparing the amplitude of the reflected wave with the amplitude threshold. The amplitude threshold determination unit 9 determines whether a reflected wave is detected within a predetermined elapsed time from a transmission of the probe wave.

[0029] The transmission time is, for example, the time at which an AC signal begins to be input from the signal generation unit 3 into the transmission circuit 2. The reception time is, for example, a time, such as time t1, that is in Fig. 5 shows the point at which the amplitude of the received wave exceeds an amplitude threshold, or a time, such as time t2, which is shown in Fig. Figure 5 shows the point at which the amplitude of the received wave is at its maximum. The reception time in Fig. 5 can be a time other than the time t1 at which the amplitude of the received wave exceeds the amplitude threshold, or a time other than the time t2 at which the amplitude of the received wave is at its maximum.

[0030] The amplitude maximum detection unit 10 compares, as described below, a gradient of the amplitude of the received wave with a gradient threshold. The reception time can be the time at which the magnitude or absolute value of the gradient of the received wave's amplitude falls below the gradient threshold. The frequency determination unit 11 also compares, as described below, a frequency matching ratio with a matching ratio threshold. The reception time can be the time at which the frequency matching ratio exceeds the matching ratio threshold. The reception time can also be the time at which the matching ratio reaches its maximum.

[0031] Results of measurements taken by the amplitude threshold determination unit 9, including propagation time and similar data, are sent to the control unit 4. Maximum values ​​are also sent to the control unit 4. A maximum value can be the amplitude of the received wave, the amplitude value at the time of reception, or an amplitude value compared to the amplitude threshold. If the amplitude threshold varies over time, the amplitude threshold at the time of reception can be used as the maximum value.

[0032] The control unit 4 calculates the distance to an object based on information sent by the amplitude threshold determination unit 9 and determines whether the object is within a predefined distance of the vehicle. As a response to the result of this determination, it sends a notification to the vehicle's driver. The control unit 4 corresponds to a distance determination unit.

[0033] The amplitude maximum detection unit 10 detects maxima of the amplitude of the received wave from the amplitude waveform generated by the amplitude generation unit 7. In the present embodiment, the amplitude maximum detection unit detects maxima of the amplitude of the received wave by comparing a gradient of the amplitude of the received wave with a predetermined gradient threshold.

[0034] The amplitude waveform of the received wave varies depending on the frequency characteristics of microphone 1, transmitting circuit 2, and receiving circuit 5. The gradient threshold is determined based on a frequency characteristic of one of the units: the transmitting unit (comprising microphone 1 and transmitting circuit 2), the receiving unit (comprising microphone 1 and receiving circuit 5), or both. For example, the gradient threshold is decreased with a narrower resonance band of microphone 1 and increased with a wider resonance band.

[0035] The frequency determination unit 11 is configured to determine, based on a waveform generated by the frequency generation unit 8, whether the frequency of the received wave changes in the same way as the probe wave. The frequency determination unit 11 also serves as a matching unit for calculating the matching ratio between the frequency of the received wave and the prescribed pattern. Based on this calculated matching ratio, the frequency determination unit 11 determines whether a signal whose frequency changes over time according to the prescribed pattern is present in the received wave.

[0036] After receiving a reflected wave of the probe wave, microphone 1 outputs a signal whose frequency changes over time in the same way as the probe wave, following an inverse shift with the probe wave, or changes more slowly than that of the pulse signal. This is because microphone 1 begins to vibrate slightly at or near its resonant frequency f0 after an AC voltage is applied to it by transmitting circuit 2, and it takes time for it to vibrate at the frequency of the pulse signal. This is also because microphone 1 begins to vibrate slightly at or near its resonant frequency f0 after receiving the reflected wave of the probe wave, and it takes time for it to vibrate at the frequency of the reflected wave.

[0037] For example, if microphone 1 transmits a first probe wave containing a positive chirp signal, the positive chirp signal, whose frequency changes in the same way as the pulse signal, is preceded in a received signal of the first probe wave by a signal whose frequency increases monotonically more slowly than that of the pulse signal. Alternatively, the positive chirp signal, whose frequency changes in the same way as the pulse signal, may be preceded in a received signal of the first probe wave by a signal whose frequency decreases monotonically.

[0038] If microphone 1 transmits a second probe wave containing a negative chirp signal, the negative chirp signal, whose frequency changes in the same way as the pulse signal, is preceded in a received signal of the second probe wave by a signal whose frequency increases monotonically. Alternatively, the negative chirp signal, whose frequency changes in the same way as the pulse signal, may be preceded in a received signal of the second probe wave by a signal whose frequency decreases monotonically more slowly than that of the pulse signal.

[0039] In the present embodiment, a determination of the received wave is thus made using a chirp signal present in the received wave and a signal preceding the chirp signal. The reference wave memory 12 stores, in particular, for each chirp signal, a reference wave corresponding to the chirp signal and a signal preceding the chirp signal. The frequency determination unit 11 calculates a degree of matching between a waveform transmitted by the frequency generation unit 8 and a waveform of each reference wave stored in the reference wave memory 12 in order to determine a chirp signal present in the received wave.

[0040] In such a resonant microphone 1, signals that are identical or similar to the signals described above are output at the reception start time of the reflected wave, at the reception end time of the reflected wave, and at the reception time of noise. In order to distinguish these signals from the chirp signal corresponding to the probe wave, in the present embodiment, as described below, a determination is made with regard to the received wave using a frequency matching ratio and maxima of the amplitude.

[0041] For example, a waveform corresponding to the positive chirp signal is used as a reference waveform whose frequency increases more slowly than that of the pulse signal generated by signal generation unit 3 and then increases at the same rate of change as the pulse signal. Alternatively, a waveform is used whose frequency decreases and then increases at the same rate of change as the pulse signal.

[0042] For example, a waveform is used as the reference wave corresponding to the negative chirp signal, whose frequency decreases more slowly than that of the pulse signal generated by signal generation unit 3 and then increases at the same rate of change as the pulse signal. Alternatively, a waveform is used whose frequency increases and then decreases at the same rate of change as the pulse signal.

[0043] The reference waves, which exhibit frequency change patterns similar to those of the probe waves, are stored in the reference wave memory 12. If the degree of match between the waveform generated by the frequency generator 8 and a specific reference wave exceeds a predetermined match threshold, the frequency determination unit 11 determines that the received wave contains a chirp signal.

[0044] The frequency determination unit 11 obtains the degree of matching using a frequency offset. Fig. Section 6 presents a method for obtaining the degree of matching for the positive chirp signal by approximating a frequency R(t) of the reference wave to a frequency f r (t) of the received wave using the frequency offset Δf and evaluating a correlation between (R(t) + Δf) and f r(t) using a residual sum of squares, where R(t) is the frequency of the reference wave and Δf is the frequency offset.

[0045] If N is assumed to be the number of samples and σ a standard deviation, the residual sum of squares E is expressed as follows: E=Nσ2=∑i=1N(fri−Ri−Δf)2

[0046] The residual sum of squares E is minimized if the following equation (2) is satisfied: ∂E∂(Δf)=0

[0047] The frequency offset Δf that satisfies equation (2) is given as follows: Δf=1N∑i=1N(fri−Ri)

[0048] The residual sum of squares E is obtained by substituting this frequency offset Δf into equation (1). A degree of matching M is obtained by substituting the residual sum of squares E obtained in this way into the following equation (4): M=1σ=NE

[0049] A degree of matching M for the negative chirp signal is obtained in the same way using the reference wave for the negative chirp signal.

[0050] The frequency determination unit 11 compares the matching ratio M with a predetermined matching ratio threshold and determines that the received wave has a chirp signal if the matching ratio M is greater than the matching ratio threshold. In particular, the frequency determination unit 11 determines that the received wave has a positive chirp signal if the matching ratio M for the positive chirp signal is greater than the matching ratio threshold, and determines that the received wave has a negative chirp signal if the matching ratio M for the negative chirp signal is greater than the matching ratio threshold.

[0051] In the present embodiment, the result of a determination performed by frequency determination unit 11 is sent to the amplitude maximum detection unit 10. The amplitude maximum detection unit 10 detects amplitude maxima using the result of a determination performed by frequency determination unit 11. The result of a determination performed by the amplitude maximum detection unit 10 is then sent to the amplitude threshold determination unit 9. The amplitude threshold determination unit 9 makes a determination regarding the amplitude based on the result of a determination performed by the amplitude maximum detection unit 10.

[0052] If, in particular, the frequency determination unit 11 determines that the received wave has a signal whose frequency changes in a predetermined pattern, the amplitude maximum detection unit 10 performs amplitude maximum detection processing. The amplitude maximum detection unit 10 defines an amplitude maximum detection range, which is a time interval relative to a time at which the degree of matching, calculated by the frequency determination unit 11, exceeds the degree of matching threshold. The amplitude maximum detection unit 10 calculates a gradient of the amplitude of the received wave within the amplitude maximum detection range and determines whether the magnitude of the calculated gradient of the amplitude of the received wave is less than the gradient threshold.

[0053] If the amplitude maximum detection unit 10 determines that the magnitude of the gradient of the amplitude of the received wave is less than the gradient threshold, the amplitude threshold determination unit 9 compares the amplitude of the received wave with the amplitude threshold.

[0054] The operating modes of the object detection device are now described. When the control unit 4 sends a transmit instruction to the signal generator 3, the signal generator 3 initiates the generation of a pulse signal. The transmit circuit 2 converts the pulse signal generated by the signal generator 3 into an AC signal, and the transmit circuit 2 applies an AC voltage to the microphone 1, causing a probe wave to be transmitted from the microphone 1. In response to the transmit instruction from the control unit 4, the signal generator 3 changes the frequency of the pulse signal to be generated over time. A first probe wave or a second probe wave is thus transmitted from the microphone 1.

[0055] If the control unit 4 subsequently sends a receive instruction to the signal processing unit 6, steps S1 to S7 are carried out as described in Fig. 7 shown, executed.

[0056] In step S1, data from the received wave is updated. Specifically, the receiver circuit 5 amplifies an output voltage from microphone 1, and the A / D converter converts the amplified output voltage and outputs it to the signal processing unit 6. The signal processing unit 6 filters a signal input by the receiver circuit 5 in response to the receive instruction.

[0057] The amplitude generator unit 7 generates an amplitude waveform from the output of the signal processing unit 6. The frequency generator unit 8 generates a frequency waveform from the output of the signal processing unit 6.

[0058] The process flow proceeds from step S1 to step S2. In step S2, for each reference wave stored in the reference wave memory 12, a determination is made as to whether the frequency waveform of the received wave and the waveform of the reference wave match. For each reference wave stored in the reference wave memory 12, the frequency determination unit 11 specifically compares the frequency waveform generated by the frequency generation unit 8 and the waveform of the reference wave to calculate a degree of match and determine whether the degree of match is greater than the degree of match threshold. If the frequency determination unit 11 determines that the degree of match is greater than the degree of match threshold, the process flow proceeds to step S3.If the frequency determination unit 11 determines that the degree of matching is equal to or less than the degree of matching threshold, the process flow proceeds to step S1.

[0059] In step S3, the amplitude maximum acquisition unit captures 10 amplitude maxima from the amplitude waveform generated by the amplitude generation unit 7. In step S3, an amplitude maximum acquisition process is performed, as described in Fig. As shown in Figure 8, the process is carried out. That is, the amplitude maximum detection unit 10 performs a normalization of the amplitude waveform in step S31, calculates a gradient of the amplitude in step S32, and determines in step S33 whether the gradient is within a predefined range. If the amplitude maximum detection unit 10 determines that the gradient is within the predefined range, it then determines that the amplitude of the received wave is at or near its maximum, which is a local maximum of the amplitude of the received wave. The process flow then proceeds to step S4. If the amplitude maximum detection unit 10 determines that the gradient is not within the predefined range, the process flow returns to step S1.

[0060] In the present embodiment, the amplitude maximum detection unit 10 performs, in particular, a normalization of the amplitude waveform in step S31. An amplitude maximum detection range is defined relative to a time at which the frequency waveform of the received wave and the waveform of the reference wave match in step S2. Amplitude values ​​sampled within the amplitude maximum detection range are divided by a representative amplitude value to obtain normalized amplitude values. In step S32, a gradient is calculated using the normalized amplitude value, and the calculated gradient is compared with the gradient threshold in step S33.

[0061] The gradient of the reflected wave's amplitude can vary with the amplitude level. If the amplitude waveform generated by the amplitude generator unit 7 is used to determine the gradient, it is necessary to vary the gradient threshold with the reflected wave's amplitude level, which can complicate the process. Therefore, normalizing the received wave's amplitude allows the gradient to be determined using a constant gradient threshold.

[0062] A time range around a reference time at which the frequency matching degree exceeds the matching degree threshold can be used as the amplitude maximum detection range. Alternative examples include a time range that starts a predefined period before the reference time and ends at the reference time, a time range between two distinct times before the reference time, or a time range between two distinct times after the reference time. The reference time can be the time at which an execution of step S31 is initiated.

[0063] The amplitude at the reference time, the amplitude at a time a predetermined interval earlier than the reference time, or the amplitude at a time a predetermined interval later than the reference time can, for example, be used as the representative amplitude value. In another alternative embodiment, a maximum or minimum amplitude value within the amplitude maximum detection range, or the first or last sampled amplitude within the amplitude maximum detection range, can be used as the representative amplitude value. In yet another alternative embodiment, a mean or median amplitude within the amplitude maximum detection range can be used as the representative amplitude value.

[0064] In the present embodiment, the amplitude maximum detection unit 10 calculates a gradient S in step S32 according to the following equation: S = (AN -A1) / T slope , where A1 represents the first sampled amplitude within the amplitude maximum detection range, A N represents the last sampled amplitude within the amplitude maximum detection range, and T slope This represents the length of the amplitude maximum detection range. The first sampled amplitude and the last sampled amplitude are, respectively, an amplitude obtained at the start time of the amplitude maximum detection range and an amplitude obtained at the end time of the amplitude maximum detection range.

[0065] If the magnitude of the gradient S in step S33 is less than the gradient threshold, the amplitude maximum detection unit 10 determines that the gradient is within the specified range. If the magnitude of the gradient S is equal to or greater than the gradient threshold, the amplitude maximum detection unit 10 determines that the gradient is not within the specified range.

[0066] The amplitude threshold determination unit 9 determines in step S4, back in Fig. 7, whether the amplitude value of the received wave is greater than the amplitude threshold or not. If it is determined that the amplitude value of the received wave is greater than the amplitude threshold, the process flow proceeds to step S5. In step S5, the result of a determination made in step S4 is stored in the amplitude threshold determination unit 9. If it is determined that the amplitude value of the received wave is equal to or less than the amplitude threshold, the process flow proceeds to step S1.

[0067] The amplitude value compared to the amplitude threshold in step S4 is a local maximum of the received wave's amplitude within the amplitude determination range, which is a time range defined based on the reference time. If no local maximum of the received wave's amplitude exists within the amplitude determination range, the amplitude value compared to the threshold can be a maximum of the received wave's amplitude within the range. In an alternative embodiment, the amplitude value compared to the threshold in step S4 can be an average of two or more amplitude values ​​within the amplitude determination range.

[0068] The amplitude determination range can be the same as the amplitude maximum detection range or different from it. For example, the amplitude determination range can be a time range that is later than the amplitude maximum detection range.

[0069] In step S6, which follows step S5, the object detection device determines whether a measured end time has elapsed. The measured end time is a predefined period after the probe wave has been transmitted. If it is determined that the measured end time has elapsed, the process flow proceeds to step S7. If it is determined that the measured end time has not yet elapsed, the process flow returns to step S1.

[0070] In step S7, the amplitude threshold determination unit 9 sends the result of a determination of the amplitude value, the maximum value(s) of the reflected wave, the reception time of the reflected wave, and the result of a chirp signal detection to the control unit 4. The control unit 4 provides a notification or the like to the driver in response to the result of a determination of the amplitude value and the maximum value(s) of the reflected wave.

[0071] The present embodiment can provide the following advantages.

[0072] Microphone 1 transmits an initial probe wave exhibiting a positive chirp signal. If the amplitude waveform and frequency waveform are as shown in Fig. As shown in Figure 9, a chirp signal is generated by the amplitude generator unit 7 and the frequency generator unit 8, and is captured as follows.

[0073] At time t3, the reference wave for the positive chirp signal, indicated by a dash-dotted line, and the frequency waveform of the received wave match, as shown in Fig. Figure 9 shows that this fits together well. If step S2 is executed with respect to time t3, the frequency matching is consequently high, and the process flow proceeds to step S3. The amplitude graph shows that the amplitude has a small gradient at time t3, as indicated by the dashed-dotted line. Consequently, step S3 determines that amplitude maxima exist at or near time t3. The process flow then proceeds to step S4, where an object is detected as a function of the amplitude magnitude.

[0074] At time t4, the reference wave for the negative chirp signal, indicated by a dash-dotted line, and the frequency waveform of the received wave match, as shown in Fig. Figure 9 shows that the frequency matching is good. When step S2 is executed with respect to time t4, the frequency matching is consequently high, and the process flow proceeds to step S3. However, the amplitude graph shows that the gradient of the amplitude at time t4, as indicated by the dashed line, is high. Consequently, step S3 determines that the amplitude is not maximal around time t4. The process flow proceeds to step S1 without the amplitude threshold determination unit 9 performing any object detection processing.

[0075] If the gradient of the amplitude is high in the present embodiment, a distance determination can thus be reversed, even if the frequency matching is high. The process flow then returns to step S1.

[0076] A chirp signal, present in the probe waveform, appears when the received signal is relatively strong. Propagation time and distance to the object can be accurately determined by detecting this chirp signal. However, in addition to this chirp signal, a frequency waveform, identical or similar to the chirp signal, may appear at the start time of reception of the reflected wave, at the end time of reception of the reflected wave, or when noise is generated. Determining the reception time based on this waveform can lead to reduced accuracy in measuring propagation time and distance to the object.

[0077] Small amplitude changes are present at or near the maxima, while large amplitude changes occur at the start and end times of reception of the reflected wave, and when noise is generated. Consequently, amplitude maxima can be detected from the amplitude gradient. Determining the reception time based on a high frequency matching and the presence of the amplitude at or near the maxima can improve the accuracy of propagation time and distance measurements. Furthermore, interference can be avoided, and simultaneous measurements using multiple microphones can be achieved, improving measurement reliability and accuracy. Second embodiment

[0078] A second embodiment is now described. This embodiment differs from the first in that a determination process for the received wave is modified. Therefore, only differences from the first embodiment are described.

[0079] In the present embodiment, the process flow proceeds as described in Fig. Figure 10 shows the process flow from step S1 to step S3. In step S3, the amplitude maximum detection unit 10 detects maxima of the amplitude. If the amplitude maximum detection unit 10 determines that the amplitude is at or near a maximum, the process flow proceeds from step S3 to step S2.

[0080] In the present embodiment, the frequency determination unit 11 compares the frequency matching ratio at a time when the amplitude is at or near its maximum with the matching ratio threshold in step S2. If step S2 determines that the frequency matching ratio is greater than the matching ratio threshold, the process flow proceeds from step S2 to step S4. In step S4, the amplitude threshold determination unit 9 determines whether the amplitude value of the received wave is greater than the amplitude threshold or not.

[0081] In this way, the positions of steps S2 and S3 are exchanged compared to the first embodiment. The present embodiment can provide similar advantages to the first embodiment.

[0082] In the present embodiment, the positions of steps S2 and S3 are interchanged. In an alternative embodiment, the positions of steps S2 through S4 can be interchanged. Steps S4, S2, and S3 can, for example, be executed in this order after step S1 has been executed. That is, in step S4, the amplitude of the received wave is compared with the amplitude threshold. In step S2, the frequency matching ratio is compared with the matching ratio threshold at a time when the amplitude exceeds the amplitude threshold. If the frequency matching ratio is greater than the matching ratio threshold, a determination is made in step S3 as to whether the amplitude is at or near its maximum. If it is determined that the amplitude is at or near its maximum, a notification is provided to the driver regarding the maximum value or similar information.

[0083] In another alternative embodiment, steps S4, S3, and S2 can be executed in this order after step S1. That is, in step S4, the amplitude of the received wave is compared to the amplitude threshold. In step S3, a determination is made as to whether the amplitude is at or near a maximum at the time it exceeds the amplitude threshold. If it is determined that the amplitude is at or near a maximum, the frequency matching is compared to the matching threshold in step S2. If it is determined that the frequency matching is greater than the matching threshold, a notification is provided to the driver regarding the maximum value or similar. Third embodiment

[0084] A third embodiment is now described. This embodiment differs from the first in that the method for detecting the amplitude maxima has been modified. Therefore, only differences from the first embodiment are described.

[0085] In the present embodiment, the amplitude maximum detection unit stores 10 reference amplitude waveforms for detecting maxima of the amplitude from the amplitude waveform of the received wave. As in Fig. As shown in Figure 11, the reference amplitude waveform is a waveform that matches a portion of a normalized amplitude waveform around a maximum of the amplitude of the received wave. The amplitude maximum detection unit 10 detects a maximum by comparing the amplitude waveform of the received wave with the reference amplitude waveform.

[0086] During the manufacturing of the object detection device, for example, a positive chirp signal and a negative chirp signal are sent in a situation where an object is positioned around the object detection device. Subsequently, a portion of the amplitude waveform of the received wave around a maximum is stored as a reference amplitude waveform in the amplitude maximum detection unit 10.

[0087] In step S3 of the present embodiment, the process flow proceeds as described in Fig. Figure 12 shows the process from step S31 to step S34. In step S34, the residual sum of squares between the normalized amplitude waveform and the reference amplitude waveform is calculated. An amplitude maximum capture range is specifically defined as a time interval at which an execution of step S3 is initiated, or at which the frequency matching degree exceeds the matching degree threshold. The residual sum of squares EA is calculated according to the following equation: EA=NσA2=∑i=1N(Ai'−RAi)2 where Ai represents the i-th sampled amplitude within the amplitude maximum detection range, Ai' represents the normalized amplitude, R Ai represents a part of the reference amplitude waveform that corresponds to the i-th sampled amplitude.

[0088] The process flow proceeds from step S34 to step S35. In step S35, a determination is made as to whether the residual sum of squares (EA) is greater than a predefined threshold value. If it is determined that the residual sum of squares (EA) is less than the predefined threshold value, the object detection device determines that the amplitude of the received wave is at or near a maximum. The process flow then proceeds to step S4. If, in step S35, the object detection device determines that the residual sum of squares (EA) is equal to or greater than the threshold value, the process flow proceeds to step S1.

[0089] The present embodiment, in which maxima of the amplitude are detected by comparing the amplitude waveform of the received wave and the reference amplitude waveform, can as such provide the same advantages as the first embodiment.

[0090] The method for detecting amplitude maxima according to the present embodiment can be applied to the second embodiment, which can provide the same advantages as the second embodiment. Fourth embodiment

[0091] A fourth embodiment is now described. This embodiment differs from the third in that the method for calculating the residual sum of squares is modified. Therefore, only differences from the third embodiment are described.

[0092] In the present embodiment, the amplitude maximum detection unit 10 detects, as in Fig. Figure 13 shows a maximum of the amplitude of the received wave by comparing the amplitude waveform of the received wave with the reference amplitude waveform multiplied by a scaling factor.

[0093] In step S3, which follows step S2, the amplitude maximum detection unit 10 calculates, as in Fig. 14 shown, in particular an amplitude scaling factor k A in step S36. The process flow goes from step S36 to step S34, in which the amplitude maximum detection unit 10 calculates a residual sum of squares EA according to the following equation: EA=NσA2=∑i=1N(Ai−kARAi)2

[0094] The amplitude scaling factor k A is calculated in such a way that the residual sum of squares is minimized. If the following equation (7) is satisfied, the residual sum of squares EA is minimized. The amplitude scaling factor k A is expressed by the following equation (8). ∂EA∂kA=0 kA=∑i=1NAi⋅RAi∑i=1NRAi2

[0095] The process flow proceeds from step S34 to step S35. In step S35, the same determination process is performed as in the third embodiment.

[0096] The present embodiment, in which the residual sum of squares is calculated, can as such provide the same advantages as the first embodiment, in which the amplitude of the received wave is normalized. Fifth embodiment

[0097] A fifth embodiment is now described. This embodiment differs from the first in that a receiving unit is added. Therefore, only differences from the first embodiment are described.

[0098] The object detection device of the present embodiment, as shown in Fig. Figure 15 shows not only receiver circuit 5, but also a receiver circuit 13. The output of microphone 1 is input to both circuits, receiver circuit 5 and receiver circuit 13. Like receiver circuit 5, receiver circuit 13 performs processing, such as amplification, noise reduction, and analog-to-digital conversion, on an output signal from microphone 1, with the gain of receiver circuit 13 being set lower than that of receiver circuit 5. Microphone 1 and receiver circuit 13 thus function as a low-gain receiver.

[0099] A signal generated by receiver circuit 5 and a signal generated by receiver circuit 13 are sent to signal processing unit 6. Signal processing unit 6 and amplitude generator unit 7 generate two amplitude waveforms in response to the two input signals. These two amplitude waveforms are sent to amplitude maximum detection unit 10.

[0100] If the amplitude of the received wave is equal to or less than a predetermined value, the amplitude maximum detection unit 10 detects a maximum of the amplitude based on an amplitude waveform generated by the output signal of the microphone 1 and the receiver circuit 5. If the amplitude of the received wave is greater than a predetermined value, the amplitude maximum detection unit 10 detects a maximum of the amplitude based on an amplitude waveform generated by the output signal of the receiver circuit 13.

[0101] If the amplitude of the received wave is high, the output of the receiving unit, in a configuration with only one receiving unit (e.g., a combination of microphone 1 and receiver circuit 5), may exceed an upper limit of energy or power acceptable to the signal processing unit 6, and information regarding the amplitude waveform may be lost, potentially leading to reduced accuracy of amplitude maximum detection. If the amplitude of the received wave exceeds a predetermined value, the amplitude maximum detection unit 10, in the present embodiment where two receiving units with different gains are used as described above, detects a maximum of the amplitude based on the output of the receiving unit with lower gain, thus avoiding a reduction in the accuracy of amplitude maximum detection. Modifications

[0102] The present invention is not limited to the above embodiments, but can be appropriately modified within the scope of the claims.

[0103] In the first embodiment, for example, a chirp signal whose frequency increases linearly and a chirp signal whose frequency decreases linearly are used. As in the Fig. 16 and Fig. As shown in Figure 17, a chirp signal whose frequency increases non-linearly and a chirp signal whose frequency decreases non-linearly can be used. The frequency of an AC signal can be determined as shown in Figure 17. Fig. Figure 18 shows that the frequency of the AC signal varies logarithmically over time, while the frequency of the AC signal can be varied discretely.

[0104] As in the Fig. 19 and Fig. As shown in Figure 20, a chirp signal can be used whose frequency varies over time after a predetermined period during which the frequency remains constant. The frequency of the chirp signal can vary linearly or non-linearly. A chirp signal can also be used whose frequency varies over time before a predetermined period during which the frequency remains constant. Finally, a chirp signal whose frequency varies discretely can be used.

[0105] In an alternative embodiment to any of the above embodiments, as described in the Fig. 21 and Fig.Figure 22 shows that a chirp signal can be used whose frequency varies in a pattern such that an increase is followed by a decrease. In this pattern, for example, a linear decrease can be followed by a linear or a non-linear increase, or a non-linear decrease can be followed by a linear or a non-linear increase. A chirp signal can be used whose frequency varies in a pattern such that a decrease is followed by an increase. In this pattern, for example, a linear increase can be followed by a linear or a non-linear decrease, or a non-linear increase can be followed by a linear or a non-linear decrease.

[0106] In the first embodiment, only one microphone 1 is provided, which serves as part of a transmitter and a receiver unit. In an alternative embodiment, two microphones can be provided. One of these two microphones serves as part of a transmitter unit, and the other serves as part of a receiver unit.

[0107] In an alternative embodiment to one of the above embodiments, the gradient of the amplitude can be calculated in a different manner than in the first embodiment. For example, for each of the sampled amplitudes within the amplitude maximum detection range, a gradient between the sampled amplitude and the previously sampled amplitude can be calculated. An average of the gradients calculated within the amplitude maximum detection range can be the gradient of the amplitude for the entire amplitude maximum detection range. In a case where the amplitude maximum detection range is unchanged, the value of the gradient of the amplitude in such an alternative embodiment takes on the same value as in the first embodiment.In one such alternative embodiment, the gradient of the amplitude is calculated each time the amplitude is sampled, allowing the object detection device to easily respond to changes in the amplitude maximum detection range. In another alternative embodiment, the gradient of the amplitude can be calculated by approximating the amplitude waveform to a linear regression model using the least squares method. In yet another alternative embodiment, the amplitude maximum detection range can be changed in response to a time elapsed since the transmission time.

[0108] In an alternative embodiment to the third embodiment, either one of the elements, reference amplitude waveform and threshold for the residual sum of squares, or both can be determined based on frequency characteristics of one of the two units, transmitting unit and receiving unit, or of both.

[0109] If, in an embodiment alternative to any of the above embodiments, a chirp signal and a maximum of the amplitude are detected in steps S2 and S3 at each of the two times within a predetermined period, a distance to the object can be determined based on the amplitude of one of the received signals at the two times within the predetermined period, the magnitude of which of the amplitude gradient is smaller than that of the other of the received signals at the two times within the predetermined period.In another embodiment alternative to any of the above embodiments, if a chirp signal and a maximum of the amplitude are detected in steps S2 and S3 at each of the two times within a predetermined period, a distance to the object can be calculated based on the amplitude of one of the received signals at the two times within the predetermined period, the residual sum of squares of which is smaller than that of the other of the received signals at the two times within the predetermined period.

[0110] In the first embodiment, the normalized amplitude of the received wave is the amplitude of the received wave divided by the representative amplitude value. In an alternative embodiment, the amplitude of the received wave can be corrected in a different way. The i-th sampled amplitude value A ithe received wave can, for example, lead to A i + ΔA are corrected, where ΔA = R Ad - A d A d represents the representative amplitude value of the received wave and R Ad represents the representative amplitude value of the reference amplitude waveform. In an alternative embodiment, ΔA can be given by ΔA = (1 / N)Σ(R Ai - A i ) where Σ represents a summation with respect to i = 1 to N. In another alternative embodiment, ΔA can be given by ΔA = (1 / N)Σ(Ai - R Ai ) be given. ΔA satisfies the following equation: δE A / δ(AA) = 0, where E A = Nσ A 2 = Σ((A i + ΔA) - R Ai ) 2 In another alternative embodiment, the i-th sampled amplitude value A can be i the received wave, for example, on A i / k A to be corrected, where k A = A d / R Ad, (1 / N)Σ(A i / R Ai ), or ΣA i / ΣR Ai In other alternative embodiments, the i-th sampled amplitude value of the received wave can, for example, be set to k. A = Σ(A i R Ai ) / Σ(R Ai 2 ) be corrected k A satisfies the following equation: ∂EA / ∂k A = 0, where E A = Nσ A 2 = Σ(A i / k A - R Ai ) 2 is.

[0111] In an alternative embodiment to the fourth embodiment, k A by k A = A d / R Ad be defined. In another alternative embodiment, k A by k A = (1 / N)Σ(A i / R Ai ) be defined. In yet another alternative embodiment, k A by k A = ΣA i / ΣR Ai be defined.

[0112] In an alternative embodiment to the fourth embodiment, A i and R Ai - ΔA can be combined, where ΔA = R Ad - A d In such an alternative embodiment, ΔA can be given by ΔA = (1 / N)Σ(R Ai - Ai). In another alternative embodiment, ΔA can be defined by ΔA = (1 / N)Σ(Ai - R Ai ) be defined as ∂E A / ∂(ΔA) = 0 is satisfied, where E A = Nσ A 2 = Σ(Ai-(R Ai - ΔA)) 2 is.

Claims

[1] Object detection device installed in a vehicle for detecting an object that is outside the vehicle, comprising: a transmitting unit (1, 2) configured to transmit probe waves which are ultrasonic waves whose frequency changes over time in a predetermined pattern, a receiving unit (1, 5) configured to receive ultrasound waves, a matching calculation unit (11) configured to calculate a matching ratio between a frequency of a received wave, which is an ultrasonic wave received by the receiving unit, and the specified pattern, an amplitude maximum detection unit (10) configured to detect a maximum of the amplitude of the received wave, and a distance determination unit (4) configured to determine a distance from the vehicle to the object based on the degree of matching calculated by the degree of matching calculation unit and a result of a detection of a maximum of the amplitude of the received wave obtained by the amplitude maximum detection unit. [2] Object detection device according to claim 1, wherein the amplitude maximum detection unit is configured to detect a maximum of the amplitude of the received wave by comparing an amplitude waveform of the received wave and a predetermined reference amplitude waveform. [3] Object detection device according to claim 2, wherein the amplitude maximum detection unit is configured such that when a residual sum of squares between the amplitude waveform of the received wave and the reference amplitude waveform is below a predetermined threshold for the residual sum of squares, it detects a maximum of the amplitude of the received wave. [4] Object detection device according to claim 3, wherein the distance determination unit is configured such that, when it is determined for each of the received waves at two times within a predetermined period that the degree of matching is greater than a degree of matching threshold and a maximum of the amplitude of the received wave has been detected by the amplitude maximum detection unit, it determines a distance to the object based on the amplitude of one of the received waves at the two times within the predetermined period, the residual sum of squares of which is less than that of the other of the received waves at the two times within the predetermined period. [5] Object detection device according to claim 3 or 4, wherein one of the elements, reference amplitude waveform and residual sum of squares threshold, or both are determined on the basis of a frequency characteristic by one of the units, transmitting unit and receiving unit, or by both. [6] Object detection device according to claim 1, wherein the amplitude maximum detection unit is configured to detect a maximum of the amplitude of the received wave by comparing a gradient of the amplitude of the received wave and a predetermined gradient threshold. [7] Object detection device according to claim 6, wherein the amplitude maximum detection unit is configured such that when an amount of the gradient of the amplitude of the received wave is below the gradient threshold, it detects a maximum of the amplitude of the received wave. [8] Object detection device according to claim 7, wherein the amplitude maximum detection unit is configured to compare the gradient of the amplitude of the received wave and the gradient threshold after normalizing the amplitude of the received wave. [9] Object detection device according to claim 7 or 8, wherein the distance determination unit is configured such that, when it is determined for each of the received waves at two times within a predetermined period that the degree of matching is greater than a degree of matching threshold and the magnitude of the gradient of the amplitude of the received wave is less than the gradient threshold, it determines a distance to the object based on the amplitude of one of the received waves at the two times within the predetermined period, the magnitude of the gradient of the amplitude being less than that of the other of the received waves at the two times within the predetermined period. [10] Object detection device according to one of claims 7 to 9, wherein the gradient threshold is determined based on frequency characteristics of one of the units, transmitting unit and receiving unit, or of both. [11] Object detection device according to any one of claims 1 to 10, wherein the distance determination unit is configured such that, when it is determined that the degree of matching is greater than a predetermined degree of matching threshold and a maximum of the amplitude of the received wave has been detected by the amplitude maximum detection unit, it determines a distance to the object based on the amplitude of the received wave. [12] Object detection device according to any one of claims 1 to 11, wherein the amplitude maximum detection unit (10) is configured to detect a maximum of the amplitude of the received wave within an amplitude maximum detection range between two times before and after a time at which the degree of matching exceeds a degree of matching threshold. [13] Object detection device according to any one of claims 2 to 12, further comprising a low-gain receiving unit (1, 13) having a lower gain than the receiving unit, wherein the amplitude maximum detection unit is configured such that, when the amplitude of the received wave is greater than a predetermined value, it detects a maximum of the amplitude of the received wave based on the amplitude of the ultrasonic wave received by the low-gain receiving unit. [14] Object detection device according to any one of claims 1 to 13, wherein the frequency of the probe wave increases monotonically or decreases monotonically. [15] Object detection device according to claim 14, wherein the receiving unit is configured to output a signal after receiving a reflected wave of the probe wave, the frequency of which changes over time in the same way as the probe wave after an inverse exchange with the probe wave. [16] Object detection device according to any one of claims 1 to 15, which further comprises a reference wave memory (12) which stores at least one reference wave having a waveform whose frequency changes in a predetermined pattern, wherein the matching degree calculation unit is configured to calculate the matching degree by comparing the frequency of the at least one reference wave stored in the reference wave memory and the frequency of the received wave.

Images